The topic of this post brought Luan Carvalho who shared with me an Open Source project connecting Apache Spark to Apache Kafka Schema Registry. Initially, I wanted to exclusively focus on the project but on my way I discovered some other interesting points.
At first glance, the update operation in an arbitrary stateful application looks just like another map's put function. However, it has an impact on what happens later with the state store. In this blog post, you will see an example that can eventually help you to reduce an I/O pressure of the updates.
If you've used Apache Kafka source in Structured Streaming, you undoubtedly noticed a property called maxOffsetsPerTrigger. According to the documentation, it helps to "limit on maximum number of offsets processed per trigger interval". My initial reaction to this property was, "Cool! We can enforce idempotent processing". I was not wrong, but the blog post will show you that I wasn't entirely right either!
Even though Apache Kafka supports transactional producers, they're not present in Apache Spark Kafka sink. But despite that, is it possible to implement a transactional producer in Apache Spark Structured Streaming? You should see that at the end of this article.
State store is a critical part of any stateful Structured Streaming application. It's important to know what happens when your business logic and input data interact with it. State store metrics will provide you some key insight into this interaction. If you don't know them now, no worries, it's the topic of this blog post!
If you read my blog post, you certainly noticed that very often I get lost on the internet. Fortunately, very often it helps me write blog posts. But the internet is not the only place where I can get lost. It also happens to me to do that with Apache Spark code and one of my most recent confusions was about FileSystem and FileContext classes.
Aside from the joins presented in the previous blog post, Structured Streaming also got a few other interesting new features that I will present here.
In the previous blog post, you discovered what changed for joins in Apache Spark 3.1. If you remember the summary sentence, it was not the single join changes in this new release. Apart from them, you can also do a bit more with Structured Streaming joins!
After the introductory part, it's time to share what I learned from the custom state store implementation.
After previous introductory posts, it's time to deep delve into the state store API and implement our own custom state store.
I don't know you, but me, when I first saw the code with createTempView method, I thought it created a temporary table in the metastore. But it's not true and in this blog post, you will see why.
After previous posts about native stateful operations, it's time to focus on the one where you can define your custom stateful logic.
Streaming joins are an interesting feature that heavily uses state store. Even though I already blogged about it in the past (2018), some changes were made and also - I hope so - my explanation capacity improved.
One of the not obvious things about the watermark is how it applies on the windows. At first glance, you could think that it will filter out the records produced before the watermark value. But it's not how it works for windows.
In previous blog posts you discovered how the state store interacts with dropDuplicates and limit operators. This time you will see how it's used in aggregations.
Another stateful operation requiring the state store is drop duplicates. You can use it to deduplicate your streaming data before pushing it to the sink.
It's the second follow-up Data+AI Summit post but the first one focusing on the stateful operations and their interaction with the state store.
The main Apache Spark component enabling stateful processing is StateStoreRDD. It creates a partition-based state store instance but also triggers state-based computation.
Some time ago @ArunJijo36 mentioned me on Twitter with a question about broadcasting in Structured Streaming. If, like me at this time, you don't know what happens, I think that this article will be good for you 👊
Few months ago, before the Apache Spark 3.0 features series, you probably noticed a short series about files processing in Structured Streaming. If you enjoyed it, here is a complementary note presenting the file data source :)
